Site News  |   Monitor Keywords  |   Monitor Archive  |   Organizer  |   Account Info  |

Symbol-level adaptation method for equalizer coefficients, memory, equalizer and receiver for implementing the method

Symbol-level adaptation method for equalizer coefficients, memory, equalizer and receiver for implementing the method description/claims

The present invention relates to a symbol-level adaptation method for equalizer coefficients, a memory, an equalizer and a receiver for implementing the method.

The terminology used throughout this specification corresponds to that defined in 3GPP (Third Generation Partnership Project) standards concerning CDMA (Code Division Multiple Access) communication systems like UMTS (Universal Mobile Telecommunication System).

It is noted that in CDMA communication systems, spreading is applied to the physical channels used to transmit data symbols from an emitter to a receiver. Spreading comprises at least a channelization operation which transforms every data symbol into a chip sequence made up of a plurality of chips, thus increasing the bandwidth of the transmitted signal. A chip is the minimal duration keying element. The number of chips per data symbol is called the spreading factor.

During the channelization operation, each data symbol from one channel is multiplied by a channelization code. Generally, a plurality of channels are simultaneously transmitted from the emitter to the receiver. Each channel is associated with its own channelization code. In orthogonal CDMA systems, the channelization codes are orthogonal. For example, OVSF (Orthogonal Variable Spreading Factor) codes are used.

In CDMA communication systems, transmission from the emitter to the receiver includes at least one pilot channel and a plurality of traffic channels. The pilot channel is used to transmit predetermined data symbols known by each receiver. These predetermined data symbols are called pilot symbols. The pilot channel can be despread by all receivers.

Each traffic channel is intended to be despread by a single receiver. Therefore, each traffic channel is spread using a channelization code known only by both the emitter and this receiver. The pilot channel, on the contrary, is spread using a channelization code known by the emitter and all receivers.

Equalizers are used in orthogonal CDMA receivers to equalize the channels received at the receiver, thus approximately restoring the orthogonality amongst the received chip sequences and reducing the interchip interferences (ICI). In other words, the equalizer corrects channel distortions at chip level.

Channel distortions vary in time. Thus, it is necessary to adapt the equalizer coefficients to track the channel changes. To this end, there are methods to adapt the equalizer coefficients according to channel distortions. The existing methods include the step of (hereinafter referred as step a)): executing an adaptive algorithm that calculates the value of the equalizer coefficient that minimizes an error between a pilot symbol estimation outputted by a despreader and the corresponding expected pilot symbol, the adaptive algorithm having a tunable parameter that determines how close the calculated coefficient value is to the optimal solution.

These existing methods are known as “symbol-level adaptation”, because the error to be minimized is the error between a despread pilot symbol and the corresponding expected pilot symbol. If, on the contrary, the error to be minimized is the error between a chip of the pilot symbol and the corresponding chip of the expected pilot symbol, the adaptation method is known as “chip-level adaptation”. The difference between symbol-level adaptation and chip-level adaptation is described in further detail in article D1:

Colin D. Frank, Eugene Visotsky and Upamanyu Madhow “Adaptive interference suppression for the downlink of a direct sequence CDMA system with long spreading sequence”; Journal of VLSI Signal Processing, vol. 30, no. 1, pp 273-291, March 2002.

Symbol-level adaptation methods have proven to be efficient. However, symbol-level adaptations can only be done at pilot symbol rate. In fact, it is necessary to wait for the reception of every chip of a pilot symbol before starting despreading this pilot chip sequence to obtain a reliable estimation of the pilot symbol from which the error can be calculated. For example, if the pilot channelization code has a spreading factor of 256, symbol adaptation can be carried out only every 256 chip intervals. Consequently, the symbol-level adaptation methods are slow in tracking fast-changing channels.

A solution to this problem has already been proposed in U.S. Pat. No. 6,175,588 in the name of Visotsky et al. More precisely, U.S. Pat. No. 6,175,588 discloses how to despread pilot symbols using a channelization code shorter than the full pilot channelization code so as to generate a pilot symbol estimation at a higher rate than the pilot symbol rate. However, the shorter pilot channelization code is not orthogonal to other simultaneously used channelization codes. As a result, the obtained pilot symbol estimation is strongly disturbed by other symbols that are simultaneously received over other channels. The reliability of this method is poor.

• Provisional Patent
• Utility Patent

• What Is a Patent?
• What Is a Trademark or Servicemark?
• What Is a Copyright?
• Patent Laws